For engineering leaders who are hiring

Turn shipped work into hired engineers.

GhostPost reads your team’s Slack, Jira, and Confluence and drafts X posts in your voice — sources shown on every draft. You approve everything. Nothing auto-posts.

Founding customer · $299/mo · free setup · 6-month contract

Slack#eng-releasescheckout p99 1.8s → 410ms 🎉
JiraPERF-312Shipped
Confluence📄 checkout latency postmortem
Sample X post drafted by GhostPost, with its sources shown.
Christopher
@mtropolis_chris · 2h

We cut checkout p99 from 1.8s to 410ms this sprint. Two engineers, two days, one super satisfying flame graph, and several N+1s later, we now have many learnings and are seeing a ~12% lift in last leg checkouts. What our process was to identify, prioritize, and fix this bug 👇

grounded in · #eng-releases · PERF-312 · checkout latency postmortem

The math

$20–40k

a recruiter's fee per engineering hire

$299/mo

GhostPost, founding rate

5–10 yrs

paid for by one hire from X

One hire pays for years of this.

Why it’s not slop

Grounded, or it doesn’t ship.

GhostPost connects to your Slack, Jira, and Confluence read-only, scoped to the channels and projects you choose.

Drafts come only from work it can point to. If it can’t be grounded, it doesn’t get drafted.

Sample X post drafted by GhostPost, with its sources shown.
Christopher
@mtropolis_chris · 2h

We cut checkout p99 from 1.8s to 410ms this sprint. Two engineers, two days, one super satisfying flame graph, and several N+1s later, we now have many learnings and are seeing a ~12% lift in last leg checkouts. What our process was to identify, prioritize, and fix this bug 👇

grounded in · #eng-releases · PERF-312 · checkout latency postmortem

The format

Strong engineering orgs already post like this.

X post by rahul.
rahul
@rahulgs · Jan 12

we built our own background coding agent at ramp: it's called inspect its powered by @opencode and @modal, and works with all the frontier models it has a cloud hosted version of vscode, chromium, and terminal. has all the tooling and skills a ramp engineer would have…

X post by DoorDash AI Research.
DoorDash AI Research
@AIatDoorDash · Jul 6

Models out of the box have trouble reviewing DoorDash's codebase, even when feeding them appropriate context. Single-pass AI reviewers only caught 30% of the real issues in our PRs. Instead, we built a multi-model (Sonnet 4.6 + Opus 4.8) code review agent that caught 53.6%…

X post by Steven Schwartz 🦅.
Steven Schwartz 🦅
@cultured · Jul 17

Whop is experiencing unprecedented growth; Daily ATHs at an astonishing YoY and MoM rate Gross transaction volume is now pushing $20M per day 10,000 is a low day for # businesses created # user signups are pushing 100,000 per day Much of our offensive growth product feature…

X post by Christopher.
Christopher
@mtropolis_chris · Jul 12

every tool that writes your posts only knows what you typed into it. that's why they all sound the same. the work actually worth posting about is sitting in slack and jira where nobody sees it.

so i'm building an ai ghostwriter that reads where your work actually happens (slack, jira, confluence), remembers it, and writes about what you really shipped in your voice.

to test it i built a fake 90 person company first. 1,678 slack messages, 60 jira tickets, and 60% of it all just garbage noise then i asked it what my team shipped last month. it pulled the right work, nailed who did what, and flagged a project that belonged to another team so it wouldn't get claimed (i never told it to do that)

Public posts by Ramp engineering, DoorDash engineering, and Whop. Shown as examples of the format — not GhostPost customers. The last card is the founder — @mtropolis_chris, building GhostPost in public.

How it works

Three steps. 1 hour a week.

01

Connect Slack, Jira, Confluence.

Read-only, one click each. Your security team can ask us anything.

SlackJiraConfluenceread-only
02

Generate content.

Everything your team shipped last week turned into tweets.

Done3last week
PERF-312checkout p99 1.8s → 410ms
DX-88onboarding: 3 weeks → 4 days
INFRA-201staging parity — 41 deploys, 0 rollbacks
03

Approve drafts.

Grounded drafts arrive in your queue. Approved posts ship on schedule. Nothing auto-posts.

Christopher@mtropolis_chris

We cut checkout p99 from 1.8s to 410ms this sprint…

grounded in · #eng-releases

The offer

Founding customer
$299/mo

Setup — $499 free for founding customers · 6-month contract, rate locked

Setup is where we build your voice model — a couple weeks of onboarding over email and one 30-minute call, so day one on the platform everything is pre-loaded.

  • A voice model that gets better the more you use it
  • It learns your company over time — writing about what you've shipped gets easier and easier
  • Grounded drafting from your connected tools
  • Your approval on every post — nothing auto-posts
  • Scheduled shipping via Typefully
  • A direct line to give me feedback and help steer the product

The trade: founding customers get the locked rate and a direct line to me. I get a monthly 20-minute feedback call and, if it works, a case study.

Prefer email? chris@ghostpost.dev

Chris, founder of GhostPostChris — founder, ghost·post

Questions

Things a skeptical eng leader would ask.